LATIDIA · Ciberseguridad

Mantenerse en la ruta de ataque: estado estructurado para las pruebas de penetración automatizadas de Long-Horizon

arXiv: 2609.07344v2Tipo de anuncio: reemplazar Resumen: Los agentes basados en el modelo de lenguaje grande (LLM) se aplican cada vez más a las tareas de ciberseguridad, como el descubrimiento de vulnerabilidades y las pruebas de penetración automatizadas. A largo plazo

La noticia

arXiv:2609.07344v2 Announce Type: replace Abstract: Large language model (LLM) based agents are increasingly applied to cybersecurity tasks such as vulnerability discovery and automated penetration testing. On long-horizon security tasks, however, such agents remain limited by context forgetting and intent drift: early critical facts and causal reasoning chains are lost over extended interactions, and the agent falls into aimless, repetitive exploration. This paper proposes Intentest, an intent-graph-guided automated penetration testing agent that externalizes long-horizon state from the LLM's context window onto a persistent fact-intent directed acyclic graph (DAG), thereby substantially reducing invalid transitions. We evaluate Intentest on automated penetration testing of web applications, a representative long-tail task in cybersecurity. In the